The UNC Resource develops forefront molecular graphics techniques and harnesses them into prototype research tools designed for biochemists studying macromolecules. The end objective is understanding the structure and function of proteins and nucleic acids, crucial to understanding disease and to designing drugs. Our Resource is the only molecular graphics group composed chiefly of computer scientists, not chemists. As such, we have special capabilities and facilities. We collaborate closely with biochemists and serve as bridge-builders between the disciplines. We maintain a trailblazer molecular graphics facility, a top-performance hardware-software configuration, continually advancing the state of the art and testing the trailblazer against real users. We welcome visiting chemists from everywhere and help them use the facility. We wildcat radical new molecular graphics ideas to the prototype stage. Winning ideas are spun off to the thriving commercial industry or into autonomous research projects. Although we cannot develop and support commercial products, we test each prototype on users who come here; then we distribute it to interested users, document it, help them install it, fix bugs, and provide telephone support. For the next five years, our main technical vision is to build graphics molecular models that the biochemist can interactively fold, twist, dock, etc., in real-time with the model maintaining visual and physics fidelity, holding all constrained lengths and angles, and continually minimizing free energy. This includes models that simultaneously work in real space and Fourier space. A second main project will be exploring new techniques for visualizing volumes, such as electron density maps and the electric fields around molecules. A whole new class of real-time direct volume visualization techniques offers new power. Other major activities will be building tools with collaborators doing drug design and de novo protein design, evaluating the suitability of new graphics engines for molecular laboratories, exploring new ways to address the molecular docking problem, continuing work on representing molecular surfaces, and adapting our high-function software to low-cost workstations.